core–shell metal zeolite composite catalysts for in situ
TRANSCRIPT
HAL Id: hal-03035156https://hal-normandie-univ.archives-ouvertes.fr/hal-03035156
Submitted on 2 Dec 2020
HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.
Core–Shell Metal Zeolite Composite Catalysts for InSitu Processing of Fischer–Tropsch Hydrocarbons to
Gasoline Type FuelsJan Přech, Debora R Strossi Pedrolo, Nilson R Marcilio, Bang Gu,
Aleksandra S Peregudova, Michal Mazur, Vitaly Ordomsky, ValentinValtchev, Andrei y Khodakov
To cite this version:Jan Přech, Debora R Strossi Pedrolo, Nilson R Marcilio, Bang Gu, Aleksandra S Peregudova, et al..Core–Shell Metal Zeolite Composite Catalysts for In Situ Processing of Fischer–Tropsch Hydrocarbonsto Gasoline Type Fuels. ACS Catalysis, American Chemical Society, 2020, 10 (4), pp.2544-2555.�10.1021/acscatal.9b04421�. �hal-03035156�
1
Revision, January 22nd, 2020
Core-Shell Metal Zeolite Composite Catalysts for In Situ Processing of
Fischer-Tropsch Hydrocarbons to Gasoline Type Fuels
Jan Přecha, Debora R. Strossi Pedrolo
b, c, Nilson R. Marcilio
c, Bang Gu
b, Aleksandra S.
Peregudovab, Michal Mazur
d, Vitaly V. Ordomsky
b, Valentin Valtchev
a and Andrei Y.
Khodakovb*
aLaboratoire Catalyse et Spectrochimie, ENSICAEN, 6 Boulevard Maréchal Juin 14050 Caen,
France
bUniv. Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR 8181 – UCCS – Unité de
Catalyse et Chimie du Solide, F-59000 Lille, France
cDepartamento de Engenharia Química, Universidade Federal do Rio Grande do Sul - UFRGS,
Porto Alegre, RS 90040-040, Brazil
dEaStCHEM School of Chemistry, University of St Andrews, St Andrews, KY16 9ST, UK
*Corresponding author: [email protected]
2
Abstract
Fischer-Tropsch synthesis has two main challenges related to direct production of gasoline fuels.
First, the chain length distribution of the products follows a broad and unselective Anderson-
Schulz–Flory distribution. Second, mostly linear hydrocarbons are formed in the Fischer-
Tropsch reaction, thus requiring isomerization while manufacturing gasoline fuels. The present
paper addresses a synthetic strategy for the preparation of hierarchical metal and zeolite
nanocomposite catalysts for direct synthesis of iso-paraffins from syngas. The nanocomposites
are synthesized in three steps. In the first step, the parent (core) zeolite is etched with an
ammonium fluoride solution. The etching creates small mesopores inside the zeolite crystals. In
the second step, the Ru nanoparticles prepared using water-in-oil microemulsion are deposed in
the mesopores of the zeolite. In the third step, a zeolite shell of MFI-type zeolites (silicalite-1 or
ZSM-5) is grown on the parent zeolite crystals coating both the etched surface and metallic
nanoparticles. Thus, the metal nanoparticles become entirely encapsulated inside the zeolite
matrix.
Most important parameters such as ruthenium content, zeolite mesoporosity and more
particularly the acidity of the catalyst shell, which affect the catalytic performance of the
synthesized nanocomposite materials in low temperature Fischer-Tropsch synthesis were
identified in this work. The higher relative amount of iso-paraffins was observed on the catalysts
containing a shell of ZSM-5. The proximity between metal and acid sites in the zeolite shell of
the nanocomposite catalysts is a crucial parameter for the design of efficient metal zeolite
bifunctional catalysts for selective synthesis of gasoline type fuels via Fischer-Tropsch synthesis,
while the acidity of the catalyst core has only a limited impact on the catalytic performance.
3
Keywords: hierarchical zeolite; core-shell structure; nanocomposite; ruthenium; Fischer-
Tropsch synthesis; bifunctional catalyst; isomerization
1. Introduction
The second-generation biofuels (gasoline, diesel, etc.) use all types of non-edible biomass, such
as lignocellulosic biomass, organic residues, wastes. The Biomass-to-Liquid (BTL)
thermochemical technology is based on the destruction of biomass by gasification and formation
of syngas. Syngas might be further converted into paraffins, olefins and alcohols by Fischer–
Tropsch (FT) synthesis. In the FT synthesis, the catalyst active phase is usually made of VIII
group transition metals; the choice of the active phase depends on various factors such as the
target product (fuels or chemicals), operating conditions and cost1, 2
.
The traditional FT technologies have two main challenges related to the direct production of
gasoline. First, the chain length distribution of the products follows the statistical Anderson-
Schulz–Flory distribution (ASF), which implies the same chain growth probability (α) for all
hydrocarbons. Therefore a specific range of the hydrocarbons cannot be produced with a
significant yield. The yield of the C5-C9 hydrocarbons (gasoline fraction) is limited to less than
50 % by the ASF polymerization statistics. Second, mostly linear hydrocarbons are formed and
thus, an isomerization of the crude FT hydrocarbons is required for production of high-octane
gasoline fuels 3.
Since the temperature for low temperature FT synthesis and isomerization are rather similar,
syngas can be also converted to iso-pararafins in a single process over bifunctional catalysts. The
selectivity to the branched hydrocarbons in direct syngas conversion can be improved by adding
4
an acid function to the FT catalysts 4,5
. The acidic function is usually provided by a zeolite, such
as ZSM-5, beta and mordenite. In the bifunctional catalysts, the concentrations, intrinsic
activities of the metal and acid sites and in particular, their localization within the catalyst, are
extremely important for the catalytic performance 6. The close proximity, “site intimacy”,
between the metal and acid sites is required to attain maximum activity and selectivity in many
reactions of bifunctional catalysis.
Several configurations of metallic and acid functions in the bifunctional catalysts and combining
FT-isomerization process have been described in the literature 7–11
. Dual-bed reactors with an
acidic zeolite downstream from a conventional FT catalyst and hybrid catalysts containing a
physical mixture of the two components have mainly been utilized for the production of
gasoline-range (C5-C12) hydrocarbons. However, the isomerization efficiency of these catalytic
systems has been insufficient. Indeed, the transformation of the formed wax by hydrocracking
over metal-free zeolites requires higher temperatures. The yield of gasoline products remains low
(Ciso/Cn<0.2) because of poor contact between the formed wax and acid sites in the zeolites.
Addition of a metal to the zeolite, usually via impregnation or ion-exchange, is often the first
step in the preparation of the composite catalyst 12
. However, metal ions partially participate in
the ion exchange with the acid sites and thus, neutralize the acidic function. In addition,
significant diffusion limitations because of extremely small pore sizes of zeolites (typically 0.5-
0.7 nm) lead to slow diffusion of hydrocarbons inside the micropores of the zeolite. The use of
mesoporous (hierarchical) zeolites for synthesis of bifunctional catalysts is a common approach
to overcome the diffusional limitations and to increase carbon monoxide conversion 13
.
Localization of metal nanoparticles inside the zeolite micropores also introduces shape-
selectivity effects for the formed hydrocarbons 14,15
. Recently, we found 16
that FT reaction
5
selectivity to short chain iso-paraffins was principally affected by the zeolite acidity, while the
selectivity to slowly diffusing long-chain hydrocarbons mostly depended on the proximity
between the metal and acid sites. The use of cobalt-zeolite nanocomposites with cobalt
nanoparticles selectively located inside the zeolite pores led to significantly higher
selectivity to the C5-C12 branched hydrocarbons in FT synthesis 17
. However, the stability
of the composite catalysts prepared by impregnation is a critical issue. In the presence of
reaction mixture and in particular water, metal nanoparticles can migrate out the zeolite
pores and agglomerate on the zeolite outer surface. This results in formation of large
metal nanoparticles on the zeolite surface and decrease in both reaction rate and
selectivity to iso-paraffins. In addition, diffusion limitations result in higher selectivity to
methane, which is a low cost product, and thus decreases the yield of the desired longer
hydrocarbons 18
. To increase efficiency of the secondary (isomerization) reactions in the
bifunctional process, Tsubaki and co-workers 19,20
and Zhu and Bollas 21
designed capsule core–
shell catalytic systems, which contain conventional FT catalyst such as Co/SiO2 or Co/Al2O3 in
the core and an acidic zeolite membrane shell on the surface of the catalyst grains.
Another approach involves using Co/SiO2 as precursor of Co-zeolite catalysts 22–24
. In principle,
the SiO2 carrier acts as a silica source for the zeolite (ZSM-5) crystallization, while the Co3O4
nanoparticles become encapsulated in the formed zeolite. The developed bifunctional catalysts
exhibit high selectivity to gasoline range of hydrocarbons with a higher yield of isomers in
comparison with hybrid and composite catalysts prepared by impregnation. Nevertheless, a thin
layer of the zeolite on the surface of metallic catalyst cannot provide high efficiency in
isomerization of the produced hydrocarbons. Increase in the zeolite layer thickness increases
diffusion limitations and thus methane production. Metal species can also be added directly into
6
a conventional zeolite synthesis gel 25
, but they often result in uncontrollable metal distribution
within the zeolite crystals. Recently, we proposed 27, 28
introduction of metal species to the
zeolites during their synthesis using metal containing carbon nanotubes as secondary hard
templates. The metal carbon nanotube templates play several roles 27
; first they serve as a replica
to create zeolites with specific fibrous morphology. Second, they markedly increase the zeolite
secondary porosity. Finally, metal carbon nanotube templates introduce the metal nanoparticles
uniformly inside the mesoporous zeolites, however, without their complete encapsulation.
In this manuscript, we propose a new strategy for the synthesis of efficient metal zeolite
bifunctional catalysts for the direct synthesis of branched gasoline type hydrocarbons from
syngas. This strategy addresses nanocasting of preformed ruthenium metallic nanoparticles in a
secondary pore system of a mesoporous zeolite. The metal nanoparticles located in the
mesopores of the zeolite are then entirely encapsulated by secondary grown zeolite structure.
This new approach offers a number of advantages. First, the separate preparation of metal
nanoparticles allows a fine control of their sizes, morphology and chemical composition. In
contrast to the in-situ synthesis of metal nanoparticles inside of the zeolites pores, the impact of
the zeolite matrix on the size and shape of the nanoparticles is suppressed. Futhermore, the
nanoparticles are stabilized in the encapsulated systems thus suppressing the particle migration
and sintering. Finally, the proximity between metal and zeolite active sites provides an
unravelled synergy between the two types of the sites in the catalyst. The zeolite performance in
hydrocarbon isomerization is mostly affected by the composition of the zeolite shell. The
prepared catalysts were characterized using a wide range of techniques, which provide
information on their morphology, structure, texture and chemical composition. Their catalytic
7
performance in the FT reaction was evaluated under high pressure in a fixed bed microreactors
using a high throughput catalytic setup.
2. Experimental
2.1.Synthesis of parent zeolites used as cores
Two different parent zeolites have been used for the preparation of the metal zeolite composite
catalysts. A sample of ZSM-5 zeolite (MFI structure) with Si/Al=21 was obtained from Süd-
Chemie in the NH4+ form. The ZSM-5 has uniform coffin shaped crystals, size about 4x2x1 µm.
Silicalite-1 (S-1, purely siliceous MFI zeolite) was synthesised as follows: 114 ml of 1 M
solution of tetrapropylammonium hydroxide (TPA-OH, Alfa Aesar) were diluted with 97.8 g of
distilled water and subsequently 65.1 g of Ludox HS-30 colloidal silica (30 wt.% of SiO2, Sigma
Aldrich) was added dropwise under vigorous stirring. The synthesis mixture with a molar
composition 100 SiO2 : 35 TPA-OH : 4000 H2O was homogenized for 1 h at room temperature
and subsequently transferred in a 0.5 l Teflon-lined autoclave with a magnetic stirrer. The
hydrothermal synthesis occurred at 175°C, with the stirring rate of 300 rpm, for 25 h. The solid
product was filtered, washed with copious amount of water, dried at ambient temperature and
finally calcined in static air at 500°C for 10 h with a temperature ramp of 1°C/min. The sample
has morphology of irregular crystals having 3-5 µm in size.
2.2.Zeolite etching
The etching of the parent zeolites follows the procedure described in Ref. 28
. A 40 wt.% solution
of NH4F was prepared from 80.0 g of NH4F (98%, Sigma-Aldrich) and 120.0 g of distilled water.
The solution was pre-heated to 50°C (+/- 2°C) in the 40 kHz ultrasonic bath in an open vessel
8
equipped with a mechanical stirrer. Then, 10 g of the parent zeolite were introduced into the
solution under stirring and ultrasound radiation. The etching occurred for 30 min (ZSM-5) or 13
min (S-1) under the above conditions. After the given time, the suspension was quickly filtered
on a Buchner funnel and washed several times with distilled water (about 500 ml all together).
The solid product was dried at 90°C and characterized. Note that NH4F solution is a contact
poison; use appropriate safety devices.
Before any further use, the etched ZSM-5 was ion-exchanged into Na+ form; using 1M solution
of NaNO3 for 4 h at room temperature. The procedure was repeated 4 times as 100 ml of solution
per 1 g of zeolite was used. After the treatment, the zeolite was filtered, washed twice with 100
ml/g of water and dried at ambient temperature. The etched S-1 was used without any further
pre-treatment. Please note that the Ru/ZSM-5 and Ru/S-1 labels correspond to the etched
zeolites.When the zeolite was non-etched , this is specified explicitly.
2.3.Preparation of Ru nanoparticles and their insertion in zeolite
Ruthenium nanoparticles were prepared following the procedure from Ref.29. Initially, two
micro-emulsions were prepared. Micro-emulsion I was prepared from 3.00 g of
hexadecyltrimethylammonium bromide (CTABr, 96%, Sigma-Aldrich), 5.00 g of 1-hexanol
(98%, Sigma-Aldrich) and 2 g of 0.5 M aqueous solution of RuCl3 (Ru content 45-55%, Sigma-
Aldrich). Micro-emulsion II was composed of 3.00 g of CTABr, 5.00 g of 1-hexanol and 2 g of
1.5 M aqueous solution of NaBH4 (96%, Sigma-Aldrich). Both micro-emulsions were
homogenized for 30 min and subsequently emulsion II was added dropwise into emulsion I
under vigorous stirring (750 rpm). The combined emulsions were stirred for another 30 min to
finish the reduction and then 500 mg of the etched zeolite (Na+ form in the case of ZSM-5) was
9
added and stirring continued for another 2 h. After this time, the emulsion was diluted with about
20 ml of ethanol (96%) filtered and the solid black product was washed with ethanol and
subsequently with water (about 200 ml of each). Finally, the product was dried at ambient
temperature and subjected to the overgrowing or calcined in static air at 480°C for 5 h with
temperature ramp 2°C/min. The Ru nanoparticles deposition on parent zeolites was performed
using the same procedure. The sample was calcined after the nanoparticle deposition.
2.4.Coating the Ru-zeolite catalysts with S-1 shell
For the preparation of S-1 shell, a coating solution with a molar composition of 12 TPA-OH :
100 SiO2 : 5800 H2O was prepared from 1M TPA-OH solution, tetraethyl orthosilicate (98%,
Sigma-Aldrich) and distilled water. After mixing all the three components, the mixture was
stirred overnight to obtain a clear homogenous solution.
This solution was combined with the non-calcined product of the insertion step (see section 2.3)
using 25 g of solution/1 g of the material after insertion. The overgrowing solution and etched
zeolite were stirred together for 2 h and then subjected to the secondary zeolite crystallization in
a tumbling Teflon-lined autoclave (50 rpm) at 150°C for 15 h. After the given time, the crude
product was filtered, washed with distilled water, dried at ambient temperature and calcined in
static air starting at 110°C for 2 h and then at 480°C for 5 h with a temperature ramp 2°C/min.
2.5.Coating the Ru-zeolite catalysts with the ZSM-5 shell
The ZSM-5 shell was crystallized from a coating solution with a molar composition of 25 TPA-
OH : 100 SiO2 : 3.33 Al(OH)3 : 5800 H2O (Si/Al=30). It was prepared from 1M TPA-OH
solution, tetraethyl orthosilicate (98%, Sigma-Aldrich), aluminum hydroxide (Aldrich) and
10
distilled water. After mixing all four components, the solution was stirred overnight to obtain
clear homogenous solution. The coating procedure was the same as described above for S-1 shell
with the exception that crystallization of the shell occurred at 140°C for 4 h. After workup of the
coating mixtures, the samples were calcined in static air starting at 110°C for 2 h and then at
480°C for 5 h with a temperature ramp 2°C/min.
2.6.Ion exchange and reduction
The calcined samples were ion-exchanged into the NH4+ form by four-stage treatment with 1 M
NH4NO3 solution (100 ml/g of zeolite, 4 h each time). Finally, the samples were converted into
H+-form by calcination at 450°C for 2 h (ramp 2°C/min). For characterization, a part of the
catalysts was reduced in a flow of hydrogen (100 ml/min) in a slowly rotating oven at 400°C for
5 h (ramp 2°C/min). The samples, which were used in the catalytic tests, were reduced in the
reactor just before running the catalytic test.
2.7.Characterization techniques
The X-ray powder diffraction patterns (XRD) were measured with a PANanalytical X’Pert PRO
diffractometer using CuKα radiation (λ = 0.15418 nm). The data were collected in continuous
mode over the 2θ range of 5-50° using a ¼° divergence slit. Scanning electron microscopy
(SEM) images were collected on a MIRA TESCAN microscope equipped with a field emission
gun. The images were collected with an acceleration voltage of 30 kV.
The high-resolution transmission electron microscopy (HRTEM) was performed two different
machines. The first one was Jeol JEM-2011 microscope operating at the accelerating voltage of
200 kV. The HRTEM images were recorded with a Gatan 794 CCD camera. The camera length,
11
sample position, and magnification were calibrated using standard gold film methods. A part of
the images was collected also using a Tecnai instrument, equipped with a LaB6 crystal, operating
at 200 kV. Prior to the analysis, the samples were dispersed by ultrasound in ethanol solution for
5 min, and a drop of solution was deposited onto a carbon membrane onto a 300 mesh-copper
grid.
Nitrogen adsorption/desorption isotherms were measured at liquid nitrogen temperature (-196°C)
using a Micromeritics Triflex instrument. Before the measurements, the samples were degassed
under vacuum of turbomolecular pump at 300°C for 6 h. The BET area was evaluated using
adsorption data in the range of a relative pressure from p/p0 = 1.1. 10-7
to p/p0 = 0.01
(corresponding to the growing part of Rouquerol BET plot). The t-plot method 30
was applied to
determine the external surface area (Sext). Micropore volume (Vmic) was determined from BJH
pore size analysis using a model for N2 on oxide surface with cylindrical geometry (part of the
Micromeritics user software). Adsorbed amount of nitrogen at p/p0 = 0.95 was used to determine
total pore volume (Vtotal).
The Si/Al ratio of the samples was determined by an ICP-OES analysis. The samples were
dissolved in a mixture of HF and Aqua Regia before the analysis. Ruthenium content was
determined from hydrogen consumption in Temperature-programmed Reduction (TPR). The
TPR experiments were performed using a Micromeritics Autochem II 2920 instrument. Before
the analysis, ruthenium was turned into RuO2 by calcination in air at 500 °C for 3 h. The TPR
was conducted in the range from 100 to 1000°C with a temperature ramp of 10°C/min. A typical
RuO2 reduction temperature was 180°C.
The concentration of acid sited was determined from pyridine titration using IR spectroscopy.
The individual samples were diluted 3 times with silica (Aldrich) to maintain sufficient
12
transparency, pressed into self-supporting wafers with a density of 8.0–12 mg/cm2 and activated
at 450°C for 5 h with a temperature ramp 2°C/min under vacuum of turbomolecular pump (final
pressure 1.0-1.5 10-6
torr). The amount of acid sites was determined from adsorption/desorption
of pyridine (Py) at 150°C. Adsorption occurred for 30 min at partial pressure 1 torr, followed by
desorption at 150°C for 15 min under vacuum of turbomolecular pump. This time is sufficient to
stabilize amount of adsorbed pyridine. The spectra were collected with a Thermo Scientific
Nicolet 6700 FT-IR spectrometer at 4 cm-1
optical resolution by collecting 128 scans for a single
spectrum. The spectra intensities were recalculated to a wafer density of 10 mg/cm2. The
concentrations of acid sites were determined from integral intensities of bands at 1454 cm−1
(Lewis) and at 1545cm−1
(Brønsted acid sites) using extinction coefficients, ε(L) = 1.28
cm/μmol, and ε(B) =1.13 cm/μmol 31
.
2.8.Catalytic testing
Carbon monoxide hydrogenation (Fisher-Tropsch reaction) was carried out on REALCAT
platform in a Flowrence® high-throughput unit (Avantium) equipped with 16 parallel milli-
fixed-bed reactors (d=2 mm) operating at the pressure of 20 bar, H2/CO= 2 molar ratio, T=220
ºC and GHSV from 4.8 -6.5 L h-1
gcat-1
. The catalyst loading was 50 mg per reactor. Prior to the
catalytic test, all the samples were activated in a flow of hydrogen at atmospheric pressure during
10 h at 400 °C. During the activation step, the temperature ramp was 3 °C/min. After the
reduction, the catalysts were cooled down to 180 °C and a flow of premixed syngas was
gradually introduced to the catalysts. When the reactor attained required pressure, the
temperature was slowly increased to the temperature of the reaction. The reaction has been
conducted for 60 h at different GHSV of syngas. Gaseous reaction products were analyzed by
13
on-line gas chromatography. The analysis of permanent gases was performed using a Molecular
Sieve column and a thermal conductivity detector. The C1-C4 hydrocarbons were separated in a
PPQ column and analyzed by a thermoconductivity detector. The C5-C12 hydrocarbons were
analyzed using a CP-Sil5 column and a flame-ionization detector. The carbon monoxide
contained 5 % of helium, which was used as an internal standard for calculating carbon
monoxide conversion. The product selectivity (S) was reported as the percentage of CO
converted into a given product and expressed on carbon basis.
Scheme 1: Synthesis of the zeolite catalysts with encapsulated ruthenium nanoparticles.
3. Results and discussion
3.1.Strategy for the synthesis of zeolite catalysts with encapsulated metal nanoparticles
The synthesis of the metal-zeolite composite catalysis with encapsulated metal nanoparticles is
composed of 3 major steps (Scheme 1). In the first step (etching), the parent (core) zeolite is
treated with a 40% NH4F solution 28
. The etching creates small mesopores inside the zeolite
crystals and sort of cups on the surface of the crystals, which accommodate metallic
nanoparticles in the second step (insertion) and serves as nucleation sites in the third step
(coating). The Ru nanoparticles were prepared using water-in-oil microemulsion following the
procedure 29
and then deposed in the mesopores of the zeolite. Ruthenium nanoparticles are
Nanocomposite with
Ru nanopartciles
Parent
zeoliteEtched
zeolite
Coated nanocomposite
with encapsulated Ru
nanopartciles
Coating (3)Insertion (2)Etching (1)
14
introduced in the second step (insertion) into the etched zeolite. In the third step (coating), the
zeolite shell crystallizes on the surface of parent crystals covering both etched surface and
metallic nanoparticles, which become trapped inside the zeolite matrix. The resulting composite
samples are denoted as shell/metal/core (e.g. ZSM-5/Ru/S-1 means the sample has a core
(parent) crystal of S-1, ruthenium nanoparticles and it was coated by a ZSM-5 shell).
In the present study, the Ru-zeolite composites with various combination of core and shell
materials (either acidic ZSM-5 or inert S-1) were prepared. In addition, the non-coated
composites Ru/ZSM-5, Ru/S-1, Ru/non-etched-ZSM-5 and Ru/non-etched-S-1 have been also
tested.
3.2.Catalyst structure
The prepared catalysts were characterized by a combination of techniques. Figure 1 presents the
SEM images of the ZSM-5/Ru/ZSM-5 and ZSM-5/Ru/S-1 catalysts at different stages of the
synthesis. The parent ZSM-5 zeolite (Figure 1 A) has regular coffin-shaped crystals sometimes
forming twins with a size of approximately 3-4 µm in the longest dimension. Parent S-1 (Figure
1 F) is composed of irregular crystals having 3-5 µm in diameter. Upon the fluoride etching, the
crystal twins are separated, leaving regular holes in the crystals and creating characteristic rough
surfaces on the planes, where the two crystals were in contact (Figure 1 B, G). These rough
surfaces are ideal places to accommodate the Ru nanoparticles. They can also act as seeds for the
crystallization of the new (coating) zeolite phase in the catalyst shell. After the insertion, some of
the nanoparticles agglomerate, while the rest is dispersed on the surface and accommodated in
the etched mesopores (cf. TEM image Ru/ZSM-5) - images collected using BSE (back-scattering
electron) detector highlights the presence of dense metallic phase (Figure 1 D,I). Upon
15
subsequent crystallization (coating), a thin layer of small ZSM-5 crystals is formed and
encapsulates the Ru nanoparticles inside the zeolitic phase (Figure 1 E, J). Note that original
shape of the parent crystals can be clearly identified by SEM and TEM through all the stages.
The S-1 shell crystallizes in slightly different morphology. The formed phase appears better
connected to the parent crystals (Figure 2) and it has more uniform surface. Moreover, formation
of some larger (200-500 nm) secondary S-1 crystals occurred, however their amount was
relatively low.
16
Figure 1: SEM images of the ZSM-5/Ru/ZSM-5 (A-E) and ZSM-5/Ru/S-1 (F-J) in the course of
the synthesis: parent ZSM-5 (A), S-1 (F); fluoride etched ZSM-5 (C) and S-1 (G); Ru/ZSM-5
and Ru/S-1 after insertion of Ru nanoparticles and calcination in the view of secondary electron
(conventional) detector (C,H) and BSE detector (D,I); final overgrown ZSM-5/Ru/ZSM-5 (E)
and ZSM-5/Ru/S-1 (J).
17
Figure 2: SEM images of ZSM-5/Ru/S-1 (S-126, left) and S-1/Ru/S-1 (S-137, right).
The catalyst synthesis was followed by powder XRD. Figure 3 shows representative XRD
patterns for ZSM-5/Ru/ZSM-5 in the final stages of the synthesis. The patterns for the other
catalysts did not differ significantly. After the Ru insertion (data not presented) and coating
(Figure 3, ZSM-5/Ru/ZSM-5 as-synthesized), the XRD patterns of the as-synthesized catalyst
contained only the patterns of MFI zeolite. No XRD peaks attributed to the ruthenium phases
were observed. This is an indirect proof that the Ru nanoparticles are too small to be detected by
XRD. The size of Co nanoparticles prepared by Subramanian 29
et al. using the same method was
approximately 5 nm. We expect therefore, the Ru nanoparticles formed in microemulsion to be
of a similar size. After calcination, broad reflections of the RuO2 phase appear in the XRD
pattern at (2θ = 28.05° and 35.1°) because the metallic ruthenium quickly oxidize to the oxide,
when exposed to oxygen at elevated temperature (during calcination). The appearance of the
lines points to the increase in the particle size due to some sintering during the calcination and
oxidation of the metal. The RuO2 particle size in the final catalysts is about 20 nm (see below).
Nevertheless, the RuO2 can be easily reduced back to Ru in hydrogen atmosphere, exhibiting
18
broad XRD lines at 2θ = 38.5°, 42.3° and 44.2°. The reduction of the catalysts was performed in-
situ after charging into the reactor. SEM (Figure 1) and TEM images (Figure 4) suggest that the
Ru diffraction lines are mostly relevant to the Ru agglomerates, which can sinter to form larger
Ru crystals. XRD patterns of all the discussed catalysts are also presented in Figures S1 and S2,
Supporting Information (SI). Note that all the XRD patterns exhibit a straight baseline, which
(together with the textural properties analysis, vide infra) proves the crystallinity of the coating
shell. For comparison, the XRD pattern of a catalyst with amorphous silica shell is shown in
Figure S3, SI. In this image, a baseline elevation between 2Θ = 15° - 30° can be observed,
which is characteristic of amorphous silica.
Figure 3: Powder XRD patterns for the ZSM-5/Ru/ZSM-5 at different stages of the synthesis.
10 20 30 40 50
ZSM-5/Ru/ZSM-5 reduced
ZSM-5/Ru/ZSM-5 calcined
ZSM-5/Ru/ZSM-5 as-synthetised
Inte
ns
ity
(a
.u.)
2 Theta (°)
1000
parent ZSM-5
19
Figure 4: TEM images of the composite catalysts: (a) ZSM-5/Ru/ZSM-5, (b) ZSM-5/Ru/S-1,
(c) Ru/ZSM-5, (d) S-1/Ru/ZSM-5.
Figure 4 presents TEM images of the catalysts with a shell on the surface of the parent (core)
crystals and a fragment of parent crystal having no shell (Ru/ZSM-5). The TEM image of the
Ru/ZSM-5 catalyst, (Figure 4 c) shows the presence of small mesopores formed by the fluoride
etching. Previously we showed 28
that the mesopores created by fluoride etching might have
20
rectangular shape. The histogram of particle size distribution has (Figure S4, SI) shown that in
Ru/ZSM-5 mostly contains the particles smaller than 25 nm. It also reveals Ru nanoparticles in
these small mesopores where, in some cases; they adopt their size and shape. In the images of the
coated catalysts, particularly these with the ZSM-5 shell (Figure 4 a,d), the nanoparticles are
covered with the small ZSM-5 crystals formed during the coating. The insert in Figure 4 d
illustrates typical Ru nanoparticles (darker round spots in middle bottom of the image) size of
about 10-20 nm but even smaller nanoparticles can be found e.g. in Figure 4 a. Thus, we
conclude the size of Ru nanoparticles inside the zeolite crystals ranges between about 3 and 20
nm. The TEM image of S-1/Ru/ZSM-5 (Figure 4 b) confirms that the S-1 shell is more uniform
in comparison with the ZSM-5 overgrown catalysts and as a result, the nanoparticles are more
difficult to be observed.
The characterization data are summarized in Table 1. In general, all the catalysts exhibit similar
textural properties, since their structure is essentially the same. S-1/Ru/ZSM-5 and S-1/Ru/S-1
(the two with S-1 shell) have slightly lower BET surface area and micropore volume in
comparison with their ZSM-5 shell analogues (e.g. BET = 320 m2/g
vs. 400 m
2/g and 346 m
2/g
vs. 404 m2/g, respectively). This probably points to a lower crystallinity of the S-1 shell in
comparison with ZSM-5. The Ru content varies between 6 and 8.4 wt.% for all the catalysts,
except for the pair Ru/S-1 and Ru/ZSM-5 (11.2 wt.% and 10.9 wt.%, respectively), because these
two catalysts contain the mesopores to accommodate the Ru nanoparticles, but their Ru content
is not decreased by the subsequent coating. The Si/Al ratio (and the Al content of the catalyst)
depends on the synthesis procedure. The parent ZSM-5, as well as the etched ZSM-5, has Si/Al =
21. The ZSM-5/Ru/ZSM-5 catalyst was prepared by coating with a sol having Si/Al ratio of 30
thus causing the decrease in the Si/Al ratio to Si/Al=27. Similarly, coating with pure silica sol
21
(giving S-1 shell) resulted in further increase to Si/Al=51. Coating of the initially pure silicalite-1
Ru/S-1 with Al containing sol (Si/Al=30) provided a catalyst with overall Si/Al=168. Note that
these values represent an average over all the catalyst; the Si/Al ratio in the core part is preserved
(that means Si/Al=21 and Si/Al=∞ for ZSM-5 and S-1 core, respectively).
Table 1: Composition, textural properties and concentration of acid sites in metal zeolite
composite catalysts.
Catalyst
Shell/Me/Core
Rua,
wt.%
Si/Alb BET
c
m2/g
Vmicc,
cm3/g
Vtotc,
cm3/g
Sext,
m2/g
Brønsted
sitesd,
mmol/g
Lewis
sitesd,
mmol/g
Ru/non etched S-1 5.9 ∞ 361 0.127 0.20 23 0 0
Ru/S-1 11.2 ∞ 371 0.131 0.20 21 0 0
Si/Ru/Si 5.2 ∞ 346 0.116 0.21 68 0 0
Ru/non-etched ZSM-5 8.4 21 375 0.130 0.19 22 0.487 0.042
Ru/ZSM-5 10.9 21 365 0.121 0.22 46 0.32 0.036
S-1/Ru/ZSM-5 5.9 51 320 0.104 0.23 49 0.168 0.03
ZSM-5/Ru/S-1 6.7 168 404 0.133 0.24 41 0.049 0.018
ZSM-5/Ru/ZSM-5 7.8 27 400 0.138 0.25 59 0.465 0.057
aRu content was determined from integral H2 consumption in a TPR experiment:
b Si/Al ratio
determined by ICP-OES;c Textural properties were determined from nitrogen sorption isotherms
collected at -196°C; d concentration of acid sites was determined from pyridine
adsorption/desorption at 150°C with FT-IR quantification of pyridinium species.
22
The acid sites are of the Brønsted character (approx. 90%), except for ZSM-5/Ru/S-1, where the
share of Lewis acid sites reaches 27%. Figure S5, SI shows IR spectra used for calculation of
the acid site concentration. In the spectra of silanol region after activation, the band at 3610 cm-1
evidences the presence of bridging Al-(OH)-Si groups (Brønsted acid sites). Only in the case of
ZSM-5/Ru/S-1, its intensity is not sufficiently high to be visible. Nevertheless, the bands at 1545
cm-1
and 1455 cm-1
, observed after pyridine adsorption/desorption at 150°C, shows the presence
of both Brønsted and Lewis acid sites, respectively, like in all other Al containing catalysts.
Figure 5: N2 sorption isotherms of the catalysts with ZSM-5 core: ZSM-5/Ru/ZSM-5 (blue), S-
1/Ru/ZSM-5 (black), Ru/ZSM-5 (red), Ru/n.e.ZSM-5 (green); empty points denote desorption.
23
Figures 5 and 6 present the nitrogen sorption isotherms, which served for the calculation of
textural properties displayed in Table 1. The isotherm of Ru/non-etched ZSM-5 is very similar
to the one of parent ZSM-5 before etching (not presented). The slight increase in the N2 uptake at
high p/p0 results from interparticle adsorption in the agglomerates of Ru nanoparticles. Ru/ZSM-
5 takes more N2 into the etched mesopores at high p/p0 (cf. TEM images). The coated ZSM-
5/Ru/ZSM-5 and S-1/Ru/ZSM-5 show a hysteresis because of intercrystalline adsorption in the
shell, which is composed of small ZSM-5 crystals. The high uptake above p/p0 = 0.9 exhibited by
S-1/Ru/ZSM-5 confirms that the shell is not 100% crystalline but there is no baseline elevation
around 2θ = 20° - 25° in the XRD pattern of this catalyst, which was observed for catalysts with
completely amorphous shell (Figure S3, SI).
Figure 6: N2 sorption isotherms of the catalysts with S-1 core: ZSM-5/Ru/S-1 (blue), S-1/Ru/S-1
(black), Ru/S-1 (red), Ru/n.e.S-1 (green); empty points denote desorption.
24
The catalysts prepared from S-1 parent exhibit an additional hysteresis at p/p0 = 0.1 – 0.3
(Figure 6). The difference of the isotherm of S-1 based catalysts with the parent non-etched S-1
and etched S-1 (Figure S6, SI) is most probably due to the presence of Ru nanoparticles. This
additional hysteresis (with another one at p/p0 = 10-4
– 10-2
) has been observed before for high
silica MFI zeolites 32,33
and it is not related to adsorption in mesopores (which occurs above
relative pressure of 0.35). The origin of these hystereses is still under debate. So far, a possible
explanation can be related to the occurrence of a phase transition between monoclinic and
orthorhombic form of the MFI framework 34,35
.
The shape of the hysteresis above p/p0 = 0.45 for ZSM-5/Ru/S-1 together with the shape of the
isotherm points to a well crystalline shell composed of small MFI crystals. For the Ru/S-1, the
high relative pressure hysteresis is missing and in the Ru/non-etched S-1 it is significantly
smaller resulting from interparticle adsorption in Ru agglomerates. The BJH pore size
distribution curves are shown in Figure S7, SI. The BJH curves clearly illustrate the formation
of mesopores upon fluoride etching and reconstruction of the material upon overgrowing. The
presence of mesoporosity in S-1 crystallized on the surface of the parent crystals (S1-Ru-ZSM-5)
also observed by SEM is possibly due to very fast crystallization so that the voids inside the
crystals were not fully reconstructed. The peak at 3.7 nm present in the BJH curves of all
samples is an artefact caused by breaking of meniscus of the liquid nitrogen at p/p0 = 0.43.
In summary, the prepared Ru-MFI composite catalysts are composed of Ru nanoparticles of 3-
20 nm (depending on preparation procedure), which are present on the surface of non-etched
parent zeolite (Ru/non-etched S-1 and Ru/non-etchedZSM-5) or in the mesopores created by the
ammonium fluoride etching (Ru/S-1 and Ru/ZSM-5). These zeolites can be covered by a layer,
25
which is composed of small secondary ZSM-5 or S-1 crystals formed during coating either with
a ZSM-5 or a S-1 shell (four catalysts having a zeolite shell). The present set of catalysts enables
to study the influence of the Ru location and influence of the acidic character of the parent (core)
zeolite and coating shell on the catalytic performance in the hydrocarbon synthesis from syngas.
Table 2. Catalytic performance data of metal-zeolite composite catalysts. Conditions:
GHSV=6500 cm3/g.h, H2/CO=2, P=20bar, T=250 °C
Catalyst
Shell/Me/Core
Ru
wt.
%
XCO,
%
RCO,
mol/g s
SCH4,
%
SC1-C12,
%
SC5-
C12,
%
SC5+, % C5-C12
iso/para
ratio
Ru/ non-etched S-1 5.9 9.0 7.25 4.2 36.8 22.7 85.9 1.6
Ru/S-1 11.2 29.5 23.79 12.0 51.7 29.9 78.2 1.8
Ru/non-etched ZSM-5 8.4 30.0 24.20
10.7 69.8 46.1 76.3 2.7
Ru/ZSM-5 10.9 49.1 39.60 12.1 88.8 62.2 73.4 3.2
S-1/Ru/S-1 4.7 8.0 6.45 10.0 47.6 26.6 79.0 1.9
Si-1/Ru/ZSM-5 5.9 23.2 18.71 12.4 61.5 36.8 75.3 1.9
ZSM-5/Ru/S-1 6.7 20.9 16.85 11.5 78.3 50.6 72.3 3.4
ZSM-5/Ru/ZSM-5 7.8 36.8 29.68 11.7 87.2 60.6 73.4 3.5
26
3.3.Catalytic performance in FT synthesis
The results of the FT catalytic evaluation of the materials are shown in Table 2, Figure 7, Table
S1 and Figures S8-S10, SI. The reaction results in the production of hydrocarbons and water.
Mostly C1-C12 hydrocarbons were produced for metal-ZSM-5 catalysts. Only negligible amounts
of carbon dioxide were detected. The dependence of carbon monoxide conversion over the
studied catalysts as function of time on stream is displayed in Figure S8, SI. No visible
deactivation was observed for all catalysts. This also suggests that no significant metal sintering
or carbon deposition occur over the studied catalysts.
Figure 7: RuTY for ruthenium zeolite nanocomposite catalysts.
0
100
200
300
400
500
600
Ru
TY, 1
0-6
s-1
27
The catalytic performance is a function of ruthenium content in the samples; higher Ru content
leads to high FT reaction rate (Figure S9, SI). Interestingly, the FT reaction rate does not
correlate with the catalyst Brönsted acidity.The catalytic performance was also compared using
ruthenium time-yield (RuTY), which was calculated as FT reaction rate normalized by the
amount of Ru atoms in the catalyst. Figure 7 shows that etching of either silicalite-1 or ZSM-5
results in a significant increase in the RuTY (e.g. S-1: 123.10-6
s-1
vs. 212.10-6
s-1
). Etching
generates a complementary mesoporous structure in the zeolite support, where the Ru
nanoparticles are subsequently deposited. Higher FT reaction rate over the etched catalysts can
be, therefore, explained by the better dispersion of Ru nanoparticles over extrnal surface and
mesopores of zeolite. Ru deposited on the etched support showed lower sintering ability during
the catalyst calcination and reduction even in case, when it is not coated.
The selectivity of FT synthesis over Ru zeolite composite catalysts depends on both porosity,
concentration and localization of acid sites within the catalyst. The Ru catalysts supported on S-1
(Ru/non-etched S-1, Ru/S-1, S-1/Ru/S-1) showed the highest selectivity to the C5+ hydrocarbons
(85.9%, 78.2% and 79%, respectively) and lowest ratio of iso- to normal hydrocarbons among all
the studied samples (1.6, 1.8, 1.9, respectively, Table 2). At the same time, the silicalite-1
etching and coating of the etched sample with a silicalite-1 layer (S-1/Ru/S-1) result in higher
selectivity to methane and lighter hydrocarbons, some decrease in the C5+ selectivity and higher
CO conversion in comparison with Ru/non etched S-1. The effect can be again due to the
localization of smaller Ru nanoparticles within the zeolite mesopores in Ru/S-1. In addition, the
S-1/Ru/S-1 provided lower CO conversion in comparison with Ru/S-1 not only becase of lower
Ru content (Ru/S-1 11.2 wt.% vs. S-1/Ru/S-1 4.7 wt.% = ratio 2.4:1) but also because of creating
additional diffusional limitation due the catalyst coating with silicalite-1 layer in the S-1/Ru/S-1
28
sample (CO conversion 29.6% vs. 8% = ratio 3.7:1). These results are consistent with previous
reports 8,36,37
. The hydrocarbon distribution curves for coated S-1/Ru/S1, S-1/Ru/ZSM-5,
ZSM/Ru/S-1 and ZSM-5/Ru/ZSM-5 catalyst are displayed in Figure S10, SI. They also clearly
show high fraction of iso-paraffins in the catalysts containing ZSM-5 in the shell (ZSM-5/Ru/S-1
and ZSM-5/Ru/ZSM-5).
The Ru catalysts supported on both non-etched and etched ZSM-5 zeolite exhibit higher
selectivity to shorter hydrocarbons in comparison with their S-1 based counterparts (e.g. SC5-C12
over Ru/S-1 29.9% vs. Ru/ZSM-5 62.2%) and at the same time higher ratio of iso- to normal
hydrocarbons in the C5-C12 hydrocarbons range is observed for these pairs (e.g. Ru/S-1 1.8 vs.
Ru/ZSM-5 3.2, Table 2). The observed phenomena are due to the isomerization and cracking of
the hydrocarbons, produced during FT synthesis, over the ZSM-5 zeolite acid sites.
Interesting results are observed for the Ru-zeolite catalysts coated with a layer of silicalite-1 or
ZSM-5 zeolite. Coating of ZSM-5 zeolites containing Ru nanoparticles with silicalite-1 (S-
1/Ru/ZSM-5) results in the selectivity patterns similar to that observed over the silicalite-1 based
samples such as Ru/S-1, Ru/non-etched S-1 or S-1/Ru/S-1 catalysts. This suggests that the
presence of the silicalite-1 shell on the top of any of the Ru zeolite catalysts significantly reduces
the isomerization activity. Coating with silicalite results in the decrease in the concentration of
the Brønsted acid sites in the proximity to Ru nanoparticles and thus, results in lower selectivity
to iso-paraffins.
Somewhat different results were obtained over the catalysts containing the shell constituted by
the ZSM-5 zeolite (regardless the core). Both ZSM-5/Ru/ZSM-5 and ZSM-5/Ru/S-1 catalysts
exhibited rather similar selectivity to methane (SCH4 = 11.5 %), to C5 and longer hydrocarbons
(SC5+ = 72.3 %) and iso/n ratio (3.4) and, together with Ru/ZSM-5 without shell, they provided
29
the highest selectivity for the gasoline hydrocarbons of all the investigated catalysts (SC5-12
60.6%, 50.6% and 62.2%, respectively). Thus, the catalysts which contain the ZSM-5 zeolite in
the shell, showed the highest isomerization activity and respectively highest ratio of the iso- to
normal hydrocarbons. This is indicative of the important and probably determining role of the
acid sites located in the shell on the selectivity patterns.
The catalytic data were also measured at higher temperature (270°C, Table S1, SI). They show
slightly lower selectivity to C5+ hydrocarbons and further increase in the ratio of iso- to normal
paraffins over the catalysts containing the acid ZSM-5 zeolite in the catalyst shell.
3.4.Roles of core, shell and porosity in the Ru-zeolite nanocomposite catalysts in the
synthesis of iso-paraffins from syngas
Our results indicate that the zeolite mesoporosity has an essential influence on the performance
of ruthenium zeolite composite catalysts. The most noticeable effect is higher FT reaction rate
observed at the same Ru content over the mesoporous samples (Figure 7). In agreement with
previous reports 13,18
, the presence of mesopores facilitates diffusion of carbon monoxide and
hydrocarbons within the catalyst pellets. The presence of mesopores reduces the effective H2/CO
ratio in the proximity to the active sites and lowers methane selectivity. Note that synthesis of
iso-paraffins from syngas over Ru zeolite composite catalysts involves both metal and acid sites.
First, the metal sites are required for CO hydrogenation to hydrocarbons via FT reaction. FT
reaction involves adsorption and dissociation of carbon monoxide and hydrogen on the metal
sites followed by surface polymerization of adsorbed CHx fragments, their hydrogenation, and
desorption 1,18
. FT synthesis results in the production of mostly linear hydrocarbons. Formation
30
of iso-paraffins under FT reaction conditions demands hydrocarbon isomerization. Isomerization
of paraffins on zeolite in the absence of metals requires high temperatures and often coincides
with major cracking. It is not likely, therefore, that paraffin isomerization solely catalyzed by
zeolite acid sites may occur under the conditions of FT synthesis. Taking into account the
temperatures typically used for low temperature FT synthesis (190-240°C), the hydrocarbon
isomerization under these conditions should also involve metal sites. Thus, the role of ruthenium
metal sites could also be related to dehydrogenation of FT hydrocarbons providing olefins for
isomerization. In this case, olefin isomerization occur on the acid sites associated with the
zeolite. Both paraffins and olefins are primary products of FT synthesis over Ru catalysts. While
olefin isomerization does not require metal sites, paraffin isomerization takes place on both metal
and acid sites. It can be favored therefore by their proximity.
The catalysts in this paper were prepared by insertion of Ru nanoparticles on and into ZSM-5 or
silicalite-1 zeolites. The inserted nanoparticles are preferentially located on the outer surface of
the zeolite particles. Generation of additional mesoporosity by etching results in partial
encapsulation of these nanoparticles and their localization in the mesopores in the subsurface
layer of the zeolite. Subsequent coating (overgrowing) introduces the silicalite-1 or ZSM-5
zeolite shell in direct interaction with Ru nanoparticles. Our characterization results are
indicative of the stability of Ru/ZSM-5 and Ru/S-1 during the coating process. Indeed, the
textural properties of the zeolite (apparent surface area, microporous volume) are only slightly
affected by the coating (Table 1, Figures 5 and 6). The XRD patterns of ZSM-5 and S-1 remain
unchanged after the coating (see Figure 2).
The catalytic results suggest that in the coated samples, ruthenium nanoparticles interact more
strongly with the coating materials (silicalite-1 or ZSM-5) than with the catalyst core. In
31
addition, any formed hydrocarbons are forced to pass through the layer of zeolite phase, where
they undergo isomerization/cracking. This explains the observed determining effect of the
catalyst shell on the reaction selectivity. The isomerization activity of the catalyst is much more
significantly affected by the shell of the composite catalysts than by the core. The isomerization
of olefins takes place mostly in the very close proximity of the ruthenium nanoparticles, which
are preferentially localized on the outer surface of the core crystals, or in the sub-surface layer
(in the case of etched samples).
Our results suggest that the presence of Brønsted acid sites in the close proximity to ruthenium
nanoparticles is indispensable in order to attain higher ratio of olefin to pararffins. Higher
concentration of Brønsted acid sites in the proximity of ruthenium nanoparticles can be attained
by creating core-shell structure with the shell formed by ZSM-5 zeolite. The presence of
Brønsted acid sites in this shell is extremely important for higher selectivity to iso-paraffins.
This shell is even more important than the catalyst core. For example, the catalysts, which are
composed by silica core (without acid sites) and zeolite shell formed by ZSM-5 zeolite, have
shown good performance in synthesis of iso-paraffins from syngas. The conclusion about
inference of the shell and its acidity on the reaction selectivity has a fundamental significance for
the design of bifunctional catalysts.
4. Conclusion
Ruthenium-zeolite composite catalysts for FT reaction have been prepared and studied with the
aim to encapsulate the metallic nanoparticles inside the zeolite matrix to provide high “intimacy”
between the metallic and acid active sites of the catalyst.
32
Efficient encapsulation of ruthenium nanoparticles within the zeolite crystals was successfully
achieved by coating the ZSM-5 and silicalite-1 carriers containing Ru nanoparticles with a shell
MFI-type material. The synthesis procedure involves thee steps: (i) creation of mesopores in the
parent zeolite by etching with ammonium fluoride, (ii) synthesis of ruthenium nanoparticles and
their deposition over parent non-etched or etched zeolites, (iii) zeolite overgrowing (coating) of
the Ru-zeolite catalysts. In these composite materials, ruthenium nanoparticles are covered a thin
layer of a zeolite shell.
The catalytic performance of the synthesized materials primarily depends on the ruthenium
content, the presence of secondary mesopores and the composition and acidity of the catalyst
shell. The reaction rates increase with higher Ru content. The diffusion limitations in the
composite catalysts are reduced by using mesoporous ZSM-5 or silicalite-1 as core. Significantly
increased relative amount of produced iso-paraffins (with respect to n-paraffins) was observed
over the catalysts containing the ZSM-5 zeolite in the shell. Isomerization of hydrocarbons
produced via FT synthesis requires the presence of both metal and acid sites. The proximity
between these sites in the zeolite shell is a crucial parameter for the design of efficient metal
zeolite bifunctional catalysts for selective synthesis of gasoline type fuels via FT synthesis. The
acidity of the catalyst core has only a limited impact on the catalytic performance.
Supporting Information
Catalytic performance measured at 270 °C; powder XRD patterns; histogram of ruthenium
particle size distribution; IR spectra of the metal-zeolite catalysts; N2 sorption isotherms; BJH
pore size distribution curves; carbon monoxide conversion as a function of time on stream; FT
reaction rate versus Ru content; hydrocarbon distribution graphs
33
Acknowledgements
The authors are grateful to Olivier Gardoll, Laurence Burylo and Joelle Thuriot for help with
TPR, XRD and XRF measurements. The authors acknowledge the financial support of the
French National Research Agency (DirectSynBioFuel project, Ref. ANR-15-CE06-0004 and
NANO4-FUT, Ref. ANR-16-CE06-0013). Chevreul Institute (FR 2638), Ministère de
l’Enseignement Supérieur, de la Recherche et de l’Innovation, Hauts-de-France Region and
FEDER are acknowledged for supporting and funding partially this work.
34
Reference
(1) Khodakov, A. Y.; Chu, W.; Fongarland, P. Advances in the Development of Novel Cobalt
Fischer-Tropsch Catalysts for Synthesis of Long-Chain Hydrocarbons and Clean Fuels.
Chem. Rev. 2007, 107, 1692–1744.
(2) Zhou, W.; Cheng, K.; Kang, J.; Zhou, C.; Subramanian, V.; Zhang, Q.; Wang, Y. New
Horizon in C1 Chemistry: Breaking the Selectivity Limitation in Transformation of
Syngas and Hydrogenation of CO2 into Hydrocarbon Chemicals and Fuels. Chem. Soc.
Rev. 2019, 48, 3193–3228.
(3) Komvokis, V. G.; Karakoulia, S.; Iliopoulou, E. F.; Papapetrou, M. C.; Vasalos, I. A.;
Lappas, A. A.; Triantafyllidis, K. S. Upgrading of Fischer–Tropsch Synthesis Bio-Waxes
via Catalytic Cracking: Effect of Acidity, Porosity and Metal Modification of Zeolitic and
Mesoporous Aluminosilicate Catalysts. Catal. Today 2012, 196, 42–55.
(4) Li, J.; He, Y.; Tan, L.; Zhang, P.; Peng, X.; Oruganti, A.; Yang, G.; Abe, H.; Wang, Y.;
Tsubaki, N. Integrated Tuneable Synthesis of Liquid Fuels via Fischer–Tropsch
Technology. Nat. Catal. 2018, 1, 787–793.
(5) Zhang, Q.; Kang, J.; Wang, Y. Development of Novel Catalysts for Fischer-Tropsch
Synthesis: Tuning the Product Selectivity. ChemCatChem 2010, 2, 1030–1058.
(6) Weisz, P. B. Polyfunctional Heterogeneous Catalysis. Adv. Catal. 1962, 13, 137–190.
(7) Martínez, A.; Valencia, S.; Murciano, R.; Cerqueira, H. S.; Costa, A. F.; S.-Aguiar, E. F.
Catalytic Behavior of Hybrid Co/SiO2-(Medium-Pore) Zeolite Catalysts during the One-
Stage Conversion of Syngas to Gasoline. Appl. Catal. A Gen. 2008, 346, 117–125.
(8) Cheng, K.; Kang, J.; Huang, S.; You, Z.; Zhang, Q.; Ding, J.; Hua, W.; Lou, Y.; Deng,
35
W.; Wang, Y. Mesoporous Beta Zeolite-Supported Ruthenium Nanoparticles for Selective
Conversion of Synthesis Gas to C 5 –C 11 Isoparaffins. ACS Catal. 2012, 2, 441–449.
(9) Sartipi, S.; Makkee, M.; Kapteijn, F.; Gascon, J. Catalysis Engineering of Bifunctional
Solids for the One-Step Synthesis of Liquid Fuels from Syngas: A Review. Catal. Sci.
Technol. 2014, 4, 893–907.
(10) Sartipi, S.; Van Dijk, J. E.; Gascon, J.; Kapteijn, F. Toward Bifunctional Catalysts for the
Direct Conversion of Syngas to Gasoline Range Hydrocarbons: H-ZSM-5 Coated Co
versus H-ZSM-5 Supported Co. Appl. Catal. A Gen. 2013, 456, 11–22.
(11) Martínez-Vargas, D. X.; Sandoval-Rangel, L.; Campuzano-Calderon, O.; Romero-Flores,
M.; Lozano, F. J.; Nigam, K. D. P.; Mendoza, A.; Montesinos-Castellanos, A. Recent
Advances in Bifunctional Catalysts for the Fischer–Tropsch Process: One-Stage
Production of Liquid Hydrocarbons from Syngas. Ind. Eng. Chem. Res. 2019, 58, 15872–
15901.
(12) Flores, C.; Batalha, N.; Marcilio, N. R.; Ordomsky, V. V.; Khodakov, A. Y. Influence of
Impregnation and Ion Exchange Sequence on Metal Localization, Acidity and Catalytic
Performance of Cobalt BEA Zeolite Catalysts in Fischer-Tropsch Synthesis.
ChemCatChem 2019, 11, 568–574.
(13) Kim, J.-C.; Lee, S.; Cho, K.; Na, K.; Lee, C.; Ryoo, R. Mesoporous MFI Zeolite
Nanosponge Supporting Cobalt Nanoparticles as a Fischer–Tropsch Catalyst with High
Yield of Branched Hydrocarbons in the Gasoline Range. ACS Catal. 2014, 4 , 3919–3927.
(14) Fraenkel, D.; Gates, B. C. Shape-Selective Fischer-Tropsch Synthesis Catalyzed by
Zeolite-Entrapped Cobalt Clusters. J. Am. Chem. Soc. 1980, 102, 2478–2480.
(15) Bao, J.; He, J.; Zhang, Y.; Yoneyama, Y.; Tsubaki, N. A Core/Shell Catalyst Produces a
36
Spatially Confined Effect and Shape Selectivity in a Consecutive Reaction. Angew.
Chemie Int. Ed. 2008, 47, 353–356.
(16) Subramanian, V.; Zholobenko, V. L.; Cheng, K.; Lancelot, C.; Heyte, S.; Thuriot, J.; Paul,
S.; Ordomsky, V. V.; Khodakov, A. Y. The Role of Steric Effects and Acidity in the
Direct Synthesis of Iso-Paraffins from Syngas on Cobalt Zeolite Catalysts. ChemCatChem
2016, 8, 380–389.
(17) Carvalho, A.; Marinova, M.; Batalha, N.; Marcilio, N. R.; Khodakov, A. Y.; Ordomsky,
V. V. Design of Nanocomposites with Cobalt Encapsulated in the Zeolite Micropores for
Selective Synthesis of Isoparaffins in Fischer-Tropsch Reaction. Catal. Sci. Technol.
2017, 7, 5019–5027.
(18) Iglesia, E.; Reyes, S. C.; Madon, R. J.; Soled, S. L. Selectivity Control and Catalyst
Design in the Fischer-Tropsch Synthesis: Sites, Pellets, and Reactors. Adv. Catal. 1993,
39, 221–302.
(19) He, J.; Liu, Z.; Yoneyama, Y.; Nishiyama, N.; Tsubaki, N. Multiple-Functional Capsule
Catalysts: A Tailor-Made Confined Reaction Environment for the Direct Synthesis of
Middle Isoparaffins from Syngas. Chem. - A Eur. J. 2006, 12, 8296–8304.
(20) He, J.; Yoneyama, Y.; Xu, B.; Nishiyama, N.; Tsubaki, N. Designing a Capsule Catalyst
and Its Application for Direct Synthesis of Middle Isoparaffins. Langmuir 2005, 21,
1699–1702.
(21) Zhu, C.; Bollas, G. M. Gasoline Selective Fischer-Tropsch Synthesis in Structured
Bifunctional Catalysts. Appl. Catal. B Environ. 2018, 235, 92–102.
(22) Liu, J.-Y.; Chen, J.-F.; Zhang, Y. Cobalt-Imbedded Zeolite Catalyst for Direct Syntheses
of Gasoline via Fischer–Tropsch Synthesis. Catal. Sci. Technol. 2013, 3, 2559-2564.
37
(23) Mazonde, B.; Cheng, S.; Zhang, G.; Javed, M.; Gao, W.; Zhang, Y.; Tao, M.; Lu, C.;
Xing, C. A Solvent-Free: In Situ Synthesis of a Hierarchical Co-Based Zeolite Catalyst
and Its Application to Tuning Fischer-Tropsch Product Selectivity. Catal. Sci. Technol.
2018, 8, 2802–2808.
(24) Liu, J.; Wang, D.; Chen, J.-F.; Zhang, Y. Cobalt Nanoparticles Imbedded into Zeolite
Crystals: A Tailor-Made Catalyst for One-Step Synthesis of Gasoline from Syngas. Int. J.
Hydrogen Energy 2016, 41, 21965–21978.
(25) Cheng, S.; Mazonde, B.; Zhang, G.; Javed, M.; Dai, P.; Cao, Y.; Tu, S.; Wu, J.; Lu, C.;
Xing, C.; et al. Co-Based MOR/ZSM-5 Composite Zeolites over a Solvent-Free Synthesis
Strategy for Improving Gasoline Selectivity. Fuel 2018, 223, 354–359.
(26) Flores, C.; Batalha, N.; Ordomsky, V. V.; Zholobenko, V. L.; Baaziz, W.; Marcilio, N. R.;
Khodakov, A. Y. Direct Production of Iso-Paraffins from Syngas over Hierarchical
Cobalt-ZSM-5 Nanocomposites Synthetized by Using Carbon Nanotubes as Sacrificial
Templates. ChemCatChem 2018, 10, 2291-2299.
(27) Flores, C.; Zholobenko, V. L.; Gu, B.; Batalha, N.; Valtchev, V.; Baaziz, W.; Ersen, O.;
Marcilio, N. R.; Ordomsky, V. V.; Khodakov, A. Y. Versatile Roles of Metal Species in
Carbon Nanotube Templates for the Synthesis of Metal–Zeolite Nanocomposite Catalysts.
ACS Appl. Nano Mater. 2019, 2, 4507–4517.
(28) Qin, Z.; Melinte, G.; Gilson, J.-P.; Jaber, M.; Bozhilov, K.; Boullay, P.; Mintova, S.;
Ersen, O.; Valtchev, V. The Mosaic Structure of Zeolite Crystals. Angew. Chemie 2016,
128, 15273–15276.
(29) Subramanian, V.; Cheng, K.; Lancelot, C.; Heyte, S.; Paul, S.; Moldovan, S.; Ersen, O.;
Marinova, M.; Ordomsky, V. V.; Khodakov, A. Y. Nanoreactors: An Efficient Tool to
38
Control the Chain-Length Distribution in Fischer-Tropsch Synthesis. ACS Catal. 2016, 6,
1785-1792.
(30) Lippens, B. Studies on Pore Systems in Catalysts V. The t Method. J. Catal. 1965, 4, 319–
323.
(31) Ngoye, F.; Lakiss, L.; Qin, Z.; Laforge, S.; Canaff, C.; Tarighi, M.; Valtchev, V.; Thomas,
K.; Vicente, A.; Gilson, J. P.; et al. Mitigating Coking during Methylcyclohexane
Transformation on HZSM-5 Zeolites with Additional Porosity. J. Catal. 2014, 320, 118–
126.
(32) Müller, U.; Unger, K. K. Sorption Studies on Large ZSM-5 Crystals: The Influence of
Aluminium Content, The Type of Exchangeable Cations and the Temperature on Nitrogen
Hysteresis Effects. Stud. Surf. Sci. Catal. 1988, 39, 101-108.
(33) Nakai, K.; Sonoda, J.; Yoshida, M.; Hakuman, M.; Naono, H. High Resolution Adsorption
Isotherms of N2 and Ar for Nonporous Silicas and MFI Zeolites. Adsorption 2007, 13,
351–356.
(34) Parra, J. B.; Ania, C. O.; Dubbeldam, D.; Vlugt, T. J. H.; Castillo, J. M.; Merkling, P. J.;
Calero, S. Unraveling the Argon Adsorption Processes in MFI-Type Zeolite. J. Phys.
Chem. C 2008, 112, 9976–9979.
(35) Cychosz, K. A.; Guillet-Nicolas, R.; García-Martínez, J.; Thommes, M. Recent Advances
in the Textural Characterization of Hierarchically Structured Nanoporous Materials.
Chem. Soc. Rev. 2017, 46, 389–414.
(36) Kang, J.; Cheng, K.; Zhang, L.; Zhang, Q.; Ding, J.; Hua, W.; Lou, Y.; Zhai, Q.; Wang, Y.
Mesoporous Zeolite-Supported Ruthenium Nanoparticles as Highly Selective Fischer-
Tropsch Catalysts for the Production of C5-C11 Isoparaffins. Angew. Chemie - Int. Ed.
39
2011, 50, 5200–5203.
(37) Zhang, Q.; Cheng, K.; Kang, J.; Deng, W.; Wang, Y. Fischer-Tropsch Catalysts for the
Production of Hydrocarbon Fuels with High Selectivity. ChemSusChem 2014, 7, 1251–
1264.